library("psych")
## Warning: package 'psych' was built under R version 3.5.3
library("skimr")
## Warning: package 'skimr' was built under R version 3.5.3
##
## Attaching package: 'skimr'
## The following object is masked from 'package:stats':
##
## filter
library("plotly")
## Warning: package 'plotly' was built under R version 3.5.3
## Loading required package: ggplot2
## Warning: package 'ggplot2' was built under R version 3.5.3
##
## Attaching package: 'ggplot2'
## The following objects are masked from 'package:psych':
##
## %+%, alpha
##
## Attaching package: 'plotly'
## The following object is masked from 'package:ggplot2':
##
## last_plot
## The following object is masked from 'package:stats':
##
## filter
## The following object is masked from 'package:graphics':
##
## layout
bank_marketing_train <- read.csv("../bank_marketing_train.csv")
# y=yes/noのデータを抽出
bank_marketing_train_y <- bank_marketing_train[bank_marketing_train$y=="yes",]
bank_marketing_train_n <- bank_marketing_train[bank_marketing_train$y=="no",]
# データ数
num_yes = dim(bank_marketing_train_y)[1]
num_no = dim(bank_marketing_train_n)[1]
# ヒストグラム(特徴が表れていそうなもの)
# 年齢(age)
pl_yes <- plot_ly(x = bank_marketing_train_y$age, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_n$age, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# => yesの方が、60以上が多い
# 職業(job)
pl_yes <- plot_ly(x = bank_marketing_train_y$job, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_n$job, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_y$job)/num_yes
## admin. blue-collar entrepreneur housemaid management
## 0.290832455 0.138040042 0.027924131 0.023182297 0.069810327
## retired self-employed services student technician
## 0.094309800 0.029504742 0.069810327 0.060590095 0.158324552
## unemployed unknown
## 0.030031612 0.007639621
summary(bank_marketing_train_n$job)/num_no
## admin. blue-collar entrepreneur housemaid management
## 0.248163483 0.235975691 0.036630159 0.026512622 0.069921197
## retired self-employed services student technician
## 0.035862161 0.033992253 0.099839722 0.016862562 0.164551890
## unemployed unknown
## 0.023640978 0.008047282
# => yesの方が、retired/studentが多く、blue-colorが少ない。特にstudentは約4倍、retiredは約3倍違いがでている
# 最終学歴(education)
pl_yes <- plot_ly(x = bank_marketing_train_y$education, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_n$education, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_y$education)/num_yes
## basic.4y basic.6y basic.9y
## 0.091938883 0.038461538 0.104320337
## high.school illiterate professional.course
## 0.223129610 0.001053741 0.127239199
## university.degree unknown
## 0.356691254 0.057165437
summary(bank_marketing_train_n$education)/num_no
## basic.4y basic.6y basic.9y
## 0.1028115400 0.0558634967 0.1534660077
## high.school illiterate professional.course
## 0.2331708294 0.0003673033 0.1262855616
## university.degree unknown
## 0.2877320689 0.0403031922
# => yesはilliterateが多い
# 連絡デバイス(contact)
pl_yes <- plot_ly(x = bank_marketing_train_y$contact, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_n$contact, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# => yesはcellularが多い
# 以前のキャンペーン結果(campaign)
pl_yes <- plot_ly(x = bank_marketing_train_y$poutcome, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_n$poutcome, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_y$poutcome)/num_yes
## failure nonexistent success
## 0.1290832 0.6817703 0.1891465
summary(bank_marketing_train_n$poutcome)/num_no# => yesはsuccessが多い(全体の割合としては2割だが、noは0.1割くらいなのでyesとnoの差はある)
## failure nonexistent success
## 0.09867103 0.88793909 0.01338988
# 以前のキャンペーンの接触回数(previous)
pl_yes <- plot_ly(x = bank_marketing_train_y$previous, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_n$previous, type="histogram", name = "no")
subplot(pl_yes, pl_no)
summary(bank_marketing_train_y$previous)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.4871 1.0000 6.0000
summary(bank_marketing_train_n$previous)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.1315 0.0000 7.0000
# => yesは平均値が大きい(yes:0.48, no:0.13)しかし、この説明変数がどれだけ有効なのかは想像つかない
-年齢:入社後の22歳ごろと退職後の60歳ごろはyesが増えそう→60歳は合っている -職業:student、unemployedはyesが少なそう→外れている。studentは逆。 -婚姻状況:divorced(離婚)はyesが少なそう→外れ。傾向なし -クレジットの支払遅延:無しはyesが多そう→外れ。傾向なし -最終学歴:調べられなかった -不動産ローンの有無:無しはyesが多そう→外れ。傾向なし -個人ローンの有無:無しはyesが多そう→外れ。傾向なし -連絡デバイス:関係なさそう→外れ。yesはcellularが多い -前回の接触からの経過日数:短い方がyesが多そう(担当者を覚えている)→外れ。傾向なし -以前のキャンペーン結果:successがyesが多そう(継続してくれるのでは)→当たり -以前のキャンペーンの接触回数:数が多い方がyesが多そう(担当者を覚えている)→当たり
## ロジスティック回帰
## 個人に紐づく、架電前に得られる説明変数のみ利用
lr<-glm(y~age+job+marital+default+education+housing+
loan+contact+day_of_week+pdays+poutcome+previous,
data=bank_marketing_train, family="binomial")
## step関数でAICを減らす
lr2 <- step(lr)
## Start: AIC=20763.61
## y ~ age + job + marital + default + education + housing + loan +
## contact + day_of_week + pdays + poutcome + previous
##
## Df Deviance AIC
## - housing 1 20690 20762
## <none> 20690 20764
## - loan 1 20693 20765
## - education 7 20712 20772
## - day_of_week 4 20707 20773
## - age 1 20702 20774
## - poutcome 2 20704 20774
## - previous 1 20706 20778
## - marital 3 20712 20780
## - pdays 1 20732 20804
## - default 2 20841 20911
## - job 11 20871 20923
## - contact 1 21013 21085
##
## Step: AIC=20762.21
## y ~ age + job + marital + default + education + loan + contact +
## day_of_week + pdays + poutcome + previous
##
## Df Deviance AIC
## <none> 20690 20762
## - loan 2 20694 20762
## - education 7 20713 20771
## - day_of_week 4 20708 20772
## - age 1 20703 20773
## - poutcome 2 20705 20773
## - previous 1 20706 20776
## - marital 3 20712 20778
## - pdays 1 20733 20803
## - default 2 20841 20909
## - job 11 20872 20922
## - contact 1 21013 21083
AIC(lr2)
## [1] 20762.21
summary(lr2)
##
## Call:
## glm(formula = y ~ age + job + marital + default + education +
## loan + contact + day_of_week + pdays + poutcome + previous,
## family = "binomial", data = bank_marketing_train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0639 -0.5021 -0.3837 -0.3010 2.8803
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.063603 0.292051 -3.642 0.000271 ***
## age 0.007940 0.002243 3.541 0.000399 ***
## jobblue-collar -0.274764 0.072316 -3.800 0.000145 ***
## jobentrepreneur -0.199841 0.112115 -1.782 0.074673 .
## jobhousemaid -0.120341 0.131644 -0.914 0.360642
## jobmanagement -0.094809 0.078278 -1.211 0.225825
## jobretired 0.612476 0.095242 6.431 1.27e-10 ***
## jobself-employed -0.156807 0.110639 -1.417 0.156401
## jobservices -0.209717 0.079214 -2.647 0.008109 **
## jobstudent 0.788676 0.104768 7.528 5.16e-14 ***
## jobtechnician -0.143732 0.064162 -2.240 0.025082 *
## jobunemployed 0.148129 0.117128 1.265 0.205986
## jobunknown -0.178919 0.227348 -0.787 0.431292
## maritalmarried 0.113629 0.062912 1.806 0.070893 .
## maritalsingle 0.298159 0.071041 4.197 2.70e-05 ***
## maritalunknown 0.126908 0.392171 0.324 0.746239
## defaultunknown -0.701074 0.060878 -11.516 < 2e-16 ***
## defaultyes -8.610256 84.476695 -0.102 0.918817
## educationbasic.6y -0.018758 0.111917 -0.168 0.866895
## educationbasic.9y -0.133104 0.086594 -1.537 0.124265
## educationhigh.school -0.062314 0.084078 -0.741 0.458606
## educationilliterate 1.003878 0.635961 1.579 0.114446
## educationprofessional.course -0.003529 0.092894 -0.038 0.969695
## educationuniversity.degree 0.067691 0.083845 0.807 0.419475
## educationunknown 0.268525 0.108378 2.478 0.013224 *
## loanunknown 0.041182 0.121540 0.339 0.734733
## loanyes -0.101550 0.052615 -1.930 0.053602 .
## contacttelephone -0.805788 0.047171 -17.082 < 2e-16 ***
## day_of_weekmon -0.141551 0.060938 -2.323 0.020186 *
## day_of_weekthu 0.059907 0.058731 1.020 0.307711
## day_of_weektue 0.054050 0.059707 0.905 0.365336
## day_of_weekwed 0.066579 0.059795 1.113 0.265510
## pdays -0.001463 0.000220 -6.649 2.95e-11 ***
## poutcomenonexistent 0.167511 0.092493 1.811 0.070130 .
## poutcomesuccess 0.800045 0.214754 3.725 0.000195 ***
## previous 0.245300 0.062066 3.952 7.74e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 23735 on 33743 degrees of freedom
## Residual deviance: 20690 on 33708 degrees of freedom
## AIC: 20762
##
## Number of Fisher Scoring iterations: 9
# => 1回目の分析で特徴が出ていたもののうち、age, job, contact, previousは影響ありそう
# => Job:retiredの条件で、データを絞ってみてみる
bank_marketing_train_job_retired <- bank_marketing_train[bank_marketing_train$job == "retired",]
summary(bank_marketing_train_job_retired)
## age job marital
## Min. :23.00 retired :1432 divorced: 289
## 1st Qu.:56.00 admin. : 0 married :1061
## Median :59.00 blue-collar : 0 single : 78
## Mean :62.15 entrepreneur: 0 unknown : 4
## 3rd Qu.:69.00 housemaid : 0
## Max. :98.00 management : 0
## (Other) : 0
## education default housing loan
## basic.4y :487 no :1115 no :645 no :1185
## university.degree :242 unknown: 317 unknown: 38 unknown: 38
## high.school :235 yes : 0 yes :749 yes : 209
## professional.course:200
## basic.9y :121
## unknown : 79
## (Other) : 68
## contact day_of_week duration campaign
## cellular :1043 fri:276 Min. : 1.0 Min. : 1.000
## telephone: 389 mon:299 1st Qu.: 116.0 1st Qu.: 1.000
## thu:254 Median : 189.0 Median : 2.000
## tue:314 Mean : 273.8 Mean : 2.443
## wed:289 3rd Qu.: 342.2 3rd Qu.: 3.000
## Max. :2093.0 Max. :42.000
##
## pdays previous poutcome emp.var.rate
## Min. : 1.0 Min. :0.0000 failure : 185 Min. :-3.4000
## 1st Qu.:999.0 1st Qu.:0.0000 nonexistent:1111 1st Qu.:-1.8000
## Median :999.0 Median :0.0000 success : 136 Median :-1.1000
## Mean :896.3 Mean :0.3191 Mean :-0.7054
## 3rd Qu.:999.0 3rd Qu.:0.0000 3rd Qu.: 1.4000
## Max. :999.0 Max. :4.0000 Max. : 1.4000
##
## cons.price.idx cons.conf.idx euribor3m nr.employed
## Min. :92.20 Min. :-50.80 Min. :0.634 Min. :4964
## 1st Qu.:92.89 1st Qu.:-42.70 1st Qu.:0.869 1st Qu.:5018
## Median :93.44 Median :-37.50 Median :1.415 Median :5099
## Mean :93.42 Mean :-38.57 Mean :2.761 Mean :5122
## 3rd Qu.:93.99 3rd Qu.:-34.80 3rd Qu.:4.959 3rd Qu.:5228
## Max. :94.77 Max. :-26.90 Max. :4.970 Max. :5228
##
## y
## no :1074
## yes: 358
##
##
##
##
##
# y=yes/noのデータを抽出してみる
bank_marketing_train_job_retired_y <- bank_marketing_train_job_retired[bank_marketing_train_job_retired$y=="yes",]
bank_marketing_train_job_retired_n <- bank_marketing_train_job_retired[bank_marketing_train_job_retired$y=="no",]
summary(bank_marketing_train_job_retired_y)
## age job marital
## Min. :33.00 retired :358 divorced: 78
## 1st Qu.:60.00 admin. : 0 married :270
## Median :68.00 blue-collar : 0 single : 10
## Mean :68.37 entrepreneur: 0 unknown : 0
## 3rd Qu.:76.00 housemaid : 0
## Max. :98.00 management : 0
## (Other) : 0
## education default housing loan
## basic.4y :148 no :330 no :149 no :297
## university.degree : 60 unknown: 28 unknown: 9 unknown: 9
## high.school : 54 yes : 0 yes :200 yes : 52
## professional.course: 45
## unknown : 28
## basic.9y : 14
## (Other) : 9
## contact day_of_week duration campaign
## cellular :319 fri:64 Min. : 63.0 Min. : 1.00
## telephone: 39 mon:61 1st Qu.: 188.2 1st Qu.: 1.00
## thu:62 Median : 311.0 Median : 1.00
## tue:88 Mean : 409.2 Mean : 1.95
## wed:83 3rd Qu.: 530.8 3rd Qu.: 2.00
## Max. :2093.0 Max. :17.00
##
## pdays previous poutcome emp.var.rate
## Min. : 2.00 Min. :0.0000 failure : 51 Min. :-3.400
## 1st Qu.: 9.25 1st Qu.:0.0000 nonexistent:209 1st Qu.:-2.900
## Median :999.00 Median :0.0000 success : 98 Median :-1.800
## Mean :713.19 Mean :0.6508 Mean :-1.939
## 3rd Qu.:999.00 3rd Qu.:1.0000 3rd Qu.:-1.700
## Max. :999.00 Max. :4.0000 Max. : 1.400
##
## cons.price.idx cons.conf.idx euribor3m nr.employed
## Min. :92.20 Min. :-50.80 Min. :0.6340 Min. :4964
## 1st Qu.:92.65 1st Qu.:-42.70 1st Qu.:0.7205 1st Qu.:5009
## Median :93.08 Median :-37.50 Median :0.8760 Median :5018
## Mean :93.24 Mean :-37.58 Mean :1.3329 Mean :5052
## 3rd Qu.:93.99 3rd Qu.:-31.40 3rd Qu.:1.3650 3rd Qu.:5099
## Max. :94.77 Max. :-26.90 Max. :4.9680 Max. :5228
##
## y
## no : 0
## yes:358
##
##
##
##
##
summary(bank_marketing_train_job_retired_n)
## age job marital
## Min. :23.00 retired :1074 divorced:211
## 1st Qu.:55.00 admin. : 0 married :791
## Median :58.00 blue-collar : 0 single : 68
## Mean :60.08 entrepreneur: 0 unknown : 4
## 3rd Qu.:63.75 housemaid : 0
## Max. :95.00 management : 0
## (Other) : 0
## education default housing loan
## basic.4y :339 no :785 no :496 no :888
## university.degree :182 unknown:289 unknown: 29 unknown: 29
## high.school :181 yes : 0 yes :549 yes :157
## professional.course:155
## basic.9y :107
## basic.6y : 58
## (Other) : 52
## contact day_of_week duration campaign
## cellular :724 fri:212 Min. : 1.0 Min. : 1.000
## telephone:350 mon:238 1st Qu.: 99.0 1st Qu.: 1.000
## thu:192 Median : 160.0 Median : 2.000
## tue:226 Mean : 228.6 Mean : 2.607
## wed:206 3rd Qu.: 282.8 3rd Qu.: 3.000
## Max. :2055.0 Max. :42.000
##
## pdays previous poutcome emp.var.rate
## Min. : 1.0 Min. :0.0000 failure :134 Min. :-3.4000
## 1st Qu.:999.0 1st Qu.:0.0000 nonexistent:902 1st Qu.:-1.8000
## Median :999.0 Median :0.0000 success : 38 Median : 1.1000
## Mean :957.4 Mean :0.2086 Mean :-0.2944
## 3rd Qu.:999.0 3rd Qu.:0.0000 3rd Qu.: 1.4000
## Max. :999.0 Max. :4.0000 Max. : 1.4000
##
## cons.price.idx cons.conf.idx euribor3m nr.employed y
## Min. :92.20 Min. :-50.8 Min. :0.635 Min. :4964 no :1074
## 1st Qu.:92.96 1st Qu.:-42.7 1st Qu.:0.993 1st Qu.:5076 yes: 0
## Median :93.44 Median :-38.3 Median :4.856 Median :5191
## Mean :93.48 Mean :-38.9 Mean :3.237 Mean :5146
## 3rd Qu.:93.99 3rd Qu.:-36.1 3rd Qu.:4.961 3rd Qu.:5228
## Max. :94.77 Max. :-26.9 Max. :4.970 Max. :5228
##
# データ数
num_retired_yes = dim(bank_marketing_train_job_retired_y)[1]
num_retired_no = dim(bank_marketing_train_job_retired_n)[1]
# 年齢
plot_ly(x = bank_marketing_train_job_retired_y$age, type="histogram")
plot_ly(x = bank_marketing_train_job_retired_n$age, type="histogram")
plot_ly(x = bank_marketing_train_job_retired$age, type="histogram", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
plot_ly(x = bank_marketing_train_job_retired$age, type="box", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
# => yesの方が、60以上が多い
# 婚姻状況
plot_ly(x = bank_marketing_train_job_retired$marital, type="histogram", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
plot_ly(x = bank_marketing_train_job_retired$marital, type="box", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$marital)/num_retired_yes
## divorced married single unknown
## 0.21787709 0.75418994 0.02793296 0.00000000
summary(bank_marketing_train_job_retired_n$marital)/num_retired_no
## divorced married single unknown
## 0.196461825 0.736499069 0.063314711 0.003724395
# => yesはsingleが少ない
# クレジットの支払遅延
plot_ly(x = bank_marketing_train_job_retired$default, type="histogram", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
plot_ly(x = bank_marketing_train_job_retired$default, type="box", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$default)/num_retired_yes
## no unknown yes
## 0.92178771 0.07821229 0.00000000
summary(bank_marketing_train_job_retired_n$default)/num_retired_no
## no unknown yes
## 0.7309125 0.2690875 0.0000000
# => yesはunknownが少なく、9割が"no"
# 最終学歴
#plot_ly(x = bank_marketing_train_job_retired$education, type="histogram", color = bank_marketing_train_job_retired$y)
#plot_ly(x = bank_marketing_train_job_retired$education, type="box", color = bank_marketing_train_job_retired$y)
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$education, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$education, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$education)/num_retired_yes
## basic.4y basic.6y basic.9y
## 0.413407821 0.019553073 0.039106145
## high.school illiterate professional.course
## 0.150837989 0.005586592 0.125698324
## university.degree unknown
## 0.167597765 0.078212291
summary(bank_marketing_train_job_retired_n$education)/num_retired_no
## basic.4y basic.6y basic.9y
## 0.3156424581 0.0540037244 0.0996275605
## high.school illiterate professional.course
## 0.1685288641 0.0009310987 0.1443202980
## university.degree unknown
## 0.1694599628 0.0474860335
# => yesはbasic.4y, illiterate(学歴が高くない)が多い
# 不動産ローンの有無
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$housing, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$housing, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$housing)/num_retired_yes
## no unknown yes
## 0.41620112 0.02513966 0.55865922
summary(bank_marketing_train_job_retired_n$housing)/num_retired_no
## no unknown yes
## 0.46182495 0.02700186 0.51117318
# => 大きな差はない(yesは少しローン有が多い)
# 個人ローンの有無
plot_ly(x = bank_marketing_train_job_retired$loan, type="histogram", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$loan, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$loan, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$loan)/num_retired_yes
## no unknown yes
## 0.82960894 0.02513966 0.14525140
summary(bank_marketing_train_job_retired_n$loan)/num_retired_no
## no unknown yes
## 0.82681564 0.02700186 0.14618250
# => 差はなさそう
# 連絡デバイス
plot_ly(x = bank_marketing_train_job_retired$contact, type="histogram", color = bank_marketing_train_job_retired$y)
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
## Warning in RColorBrewer::brewer.pal(N, "Set2"): minimal value for n is 3, returning requested palette with 3 different levels
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$contact, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$contact, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$contact)/num_retired_yes
## cellular telephone
## 0.8910615 0.1089385
summary(bank_marketing_train_job_retired_n$contact)/num_retired_no
## cellular telephone
## 0.6741155 0.3258845
# => yesはcellularが多い
# 前回の接触からの経過日数
#plot_ly(x = bank_marketing_train_job_retired$pdays, type="histogram", color = bank_marketing_train_job_retired$y)
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$pdays, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$pdays, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$pdays)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.00 9.25 999.00 713.19 999.00 999.00
summary(bank_marketing_train_job_retired_n$pdays)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.0 999.0 999.0 957.4 999.0 999.0
bank_marketing_train_job_retired_y
## age job marital education default housing loan
## 3718 58 retired married basic.4y unknown no no
## 3875 59 retired married professional.course unknown no no
## 4282 55 retired married high.school no no no
## 4447 60 retired married high.school unknown yes no
## 5561 54 retired married basic.4y unknown no no
## 5660 59 retired married professional.course no no no
## 6689 59 retired divorced university.degree no no no
## 7007 52 retired married basic.4y no no no
## 9052 59 retired married basic.9y no yes no
## 9329 55 retired married university.degree unknown yes no
## 9685 56 retired married high.school unknown yes no
## 10162 59 retired married professional.course unknown yes yes
## 10220 56 retired married basic.4y no yes no
## 11005 58 retired married basic.4y no no no
## 11383 53 retired married basic.9y unknown yes no
## 11695 57 retired married university.degree no yes yes
## 11861 54 retired married basic.4y no yes yes
## 12056 59 retired married university.degree unknown no no
## 12120 49 retired married high.school no yes no
## 12656 57 retired divorced basic.4y unknown no no
## 14096 55 retired married professional.course no yes no
## 14130 59 retired married high.school no yes no
## 14522 60 retired divorced basic.4y no yes no
## 14609 60 retired divorced professional.course no no no
## 15982 56 retired married high.school unknown yes no
## 16489 59 retired married basic.4y unknown no no
## 17004 60 retired married basic.9y unknown no no
## 17759 56 retired married basic.4y no no no
## 18092 59 retired married high.school no yes no
## 18696 60 retired married unknown no yes no
## 18905 57 retired married basic.9y no no no
## 19002 57 retired married basic.9y no yes yes
## 19063 58 retired married basic.4y no no no
## 19747 58 retired divorced basic.9y no yes no
## 20190 57 retired married university.degree no yes yes
## 21510 58 retired married high.school no no no
## 21776 50 retired divorced high.school no yes no
## 22708 70 retired divorced basic.4y no yes no
## 22783 88 retired divorced basic.4y no yes yes
## 22787 88 retired divorced basic.4y no yes no
## 22789 88 retired divorced basic.4y no yes no
## 22792 88 retired divorced basic.4y no yes no
## 22793 88 retired divorced basic.4y no yes no
## 22798 66 retired married basic.4y no yes no
## 22831 68 retired married university.degree no no no
## 22882 73 retired married university.degree no yes no
## 22931 63 retired married professional.course no yes no
## 23041 55 retired married basic.4y no yes no
## 23078 60 retired divorced basic.4y no no no
## 23173 63 retired married unknown no no no
## 23174 63 retired married unknown no yes no
## 23189 82 retired married unknown no no no
## 23208 58 retired married university.degree no unknown unknown
## 23304 73 retired divorced basic.4y no no yes
## 23323 59 retired married university.degree no no no
## 23350 61 retired married university.degree no yes no
## 23355 69 retired divorced university.degree no yes no
## 23362 70 retired married basic.4y no yes no
## 23370 70 retired married basic.4y no no no
## 23376 58 retired divorced university.degree no yes no
## 23404 66 retired married basic.4y no yes no
## 23409 67 retired married professional.course no yes no
## 23420 58 retired divorced university.degree no yes no
## 23443 63 retired married unknown no yes no
## 23530 70 retired married unknown no yes no
## 23919 71 retired single unknown no yes no
## 24043 60 retired married basic.4y unknown yes no
## 24204 68 retired married high.school no yes no
## 24293 71 retired married university.degree no no no
## 24306 75 retired divorced basic.4y no no no
## 24424 69 retired divorced basic.4y no yes no
## 24538 64 retired married university.degree no yes no
## 24544 78 retired married basic.4y no yes no
## 24545 57 retired married basic.4y no yes yes
## 24546 75 retired married basic.4y no no no
## 24550 61 retired married professional.course no unknown unknown
## 24558 65 retired married basic.4y no unknown unknown
## 24564 75 retired married basic.4y no no no
## 24568 78 retired married basic.4y no yes no
## 24570 85 retired married basic.4y unknown yes no
## 24571 64 retired married university.degree no no no
## 24589 64 retired married basic.4y no yes no
## 24599 61 retired married basic.4y no yes no
## 24600 65 retired married university.degree no no no
## 24606 58 retired single professional.course no yes no
## 24675 79 retired married basic.9y no yes no
## 24677 60 retired divorced professional.course no yes no
## 24740 59 retired married basic.4y no yes no
## 24809 69 retired married university.degree no yes no
## 24824 58 retired divorced professional.course no no no
## 24867 58 retired married basic.4y no no no
## 24903 61 retired divorced university.degree no no no
## 24980 59 retired married basic.6y no yes no
## 25111 33 retired married high.school no yes no
## 29350 81 retired divorced unknown unknown yes yes
## 29489 60 retired married university.degree no yes no
## 29630 56 retired married university.degree no no no
## 29705 59 retired married professional.course no yes yes
## 29710 65 retired married basic.9y no no no
## 29764 58 retired divorced high.school no yes no
## 29882 59 retired married professional.course no yes no
## 30093 53 retired single basic.4y no yes no
## 30122 55 retired married high.school no yes yes
## 30155 74 retired married basic.4y no yes no
## 30286 57 retired divorced basic.9y no yes no
## 30307 56 retired married high.school no yes yes
## 30405 61 retired married professional.course no no no
## 30437 59 retired married unknown no no no
## 30443 52 retired divorced university.degree no yes yes
## 30452 52 retired divorced university.degree no no no
## 30461 74 retired married high.school no yes no
## 30473 76 retired married basic.4y no yes no
## 30474 76 retired married basic.4y no yes no
## 30485 75 retired married basic.4y no no yes
## 30493 70 retired married basic.4y no yes no
## 30498 73 retired married basic.4y no no no
## 30522 76 retired married university.degree no no no
## 30588 85 retired married professional.course no no no
## 30601 80 retired married illiterate unknown yes yes
## 30637 74 retired married university.degree no yes no
## 30640 66 retired married unknown no no no
## 30724 74 retired married professional.course no yes no
## 30755 62 retired married unknown no no no
## 30776 55 retired married basic.4y no yes no
## 30778 71 retired married basic.4y no yes no
## 30780 70 retired married professional.course no yes no
## 30804 73 retired married basic.4y no no no
## 30805 67 retired married professional.course no yes yes
## 30806 80 retired married high.school no no no
## 30827 74 retired married basic.9y no yes no
## 30829 67 retired married basic.6y no no no
## 30833 61 retired married university.degree no no no
## 30850 71 retired married basic.4y unknown unknown unknown
## 30854 64 retired married high.school unknown no no
## 30859 74 retired divorced basic.4y no yes no
## 30872 74 retired married university.degree no yes yes
## 30875 64 retired married basic.4y no yes no
## 30886 66 retired married basic.6y no yes no
## 30896 71 retired single university.degree no yes no
## 30912 61 retired married basic.4y no yes yes
## 30913 61 retired married basic.4y no yes no
## 30916 87 retired divorced basic.4y no no no
## 30932 64 retired married unknown no no no
## 30963 58 retired married high.school no no no
## 30966 54 retired married professional.course unknown no no
## 30968 62 retired married professional.course no yes no
## 30970 58 retired married high.school no yes no
## 30978 42 retired divorced illiterate no no no
## 30993 58 retired divorced high.school no no no
## 31010 62 retired married professional.course no yes yes
## 31023 73 retired married basic.4y no unknown unknown
## 31053 79 retired married basic.9y no no no
## 31107 61 retired married university.degree no no no
## 31131 71 retired married basic.4y no yes no
## 31137 75 retired married unknown no no no
## 31146 58 retired married basic.4y no no no
## 31160 76 retired married basic.4y no yes no
## 31184 71 retired married unknown no yes no
## 31193 70 retired married basic.4y no no no
## 31237 70 retired divorced university.degree no yes no
## 31246 81 retired married basic.4y no yes no
## 31254 61 retired married high.school no yes no
## 31275 83 retired married professional.course no yes no
## 31279 82 retired married university.degree no yes no
## 31282 69 retired divorced professional.course no no no
## 31286 65 retired married high.school no yes no
## 31306 63 retired married high.school no no no
## 31326 77 retired married university.degree no no yes
## 31340 80 retired married basic.4y no no no
## 31363 74 retired married unknown no yes yes
## 31387 74 retired married basic.9y no yes no
## 31419 62 retired married university.degree unknown no no
## 31423 70 retired married basic.4y no yes no
## 31427 60 retired married basic.4y no no no
## 31479 56 retired married professional.course no no no
## 31495 98 retired married basic.4y unknown yes no
## 31498 98 retired married basic.4y unknown yes no
## 31525 68 retired married basic.4y no yes no
## 31544 71 retired married basic.4y no no no
## 31551 63 retired married high.school no yes no
## 31567 62 retired married high.school no no no
## 31573 55 retired married basic.4y no yes no
## 31574 80 retired divorced high.school no yes no
## 31581 72 retired married basic.6y no yes yes
## 31582 82 retired single basic.4y no yes no
## 31594 69 retired married basic.4y no yes no
## 31598 54 retired married basic.4y no no no
## 31615 69 retired married basic.4y no yes yes
## 31652 71 retired married professional.course no yes no
## 31660 79 retired married basic.4y no no yes
## 31671 71 retired married professional.course no yes yes
## 31684 73 retired married university.degree no no yes
## 31686 69 retired married basic.6y no yes no
## 31695 69 retired married high.school no no no
## 31703 76 retired single basic.4y no no no
## 31718 79 retired married basic.4y no no yes
## 31721 79 retired married basic.4y no yes no
## 31734 72 retired married basic.4y no yes no
## 31738 72 retired divorced basic.6y no yes no
## 31756 65 retired single university.degree no no no
## 31763 67 retired married basic.4y no yes no
## 31774 66 retired married basic.4y no yes yes
## 31790 80 retired divorced unknown no yes yes
## 31793 65 retired married university.degree no no no
## 31800 77 retired married unknown no yes no
## 31805 78 retired divorced basic.4y no yes no
## 31811 61 retired married basic.4y no yes no
## 31814 64 retired married high.school no no no
## 31815 67 retired married basic.4y no no no
## 31840 72 retired married high.school no no no
## 31842 77 retired divorced basic.4y no yes no
## 31847 68 retired married high.school no no no
## 31853 82 retired divorced basic.4y no yes no
## 31859 66 retired single basic.4y no yes no
## 31861 83 retired divorced basic.4y no no no
## 31872 56 retired married university.degree no yes no
## 31888 61 retired married basic.4y no yes no
## 31889 58 retired married basic.4y no yes no
## 31890 56 retired married university.degree no no no
## 31901 78 retired married professional.course no no no
## 31905 73 retired married university.degree no no no
## 31923 83 retired married high.school no no no
## 31926 60 retired married high.school no no no
## 31931 60 retired married university.degree no no no
## 31964 86 retired married professional.course no no no
## 31972 56 retired married basic.4y no no no
## 31980 81 retired divorced basic.4y no yes no
## 31989 85 retired divorced basic.4y unknown yes no
## 31996 64 retired married high.school no yes no
## 31997 70 retired married basic.4y no no no
## 32003 76 retired married university.degree no yes yes
## 32005 76 retired married university.degree no no yes
## 32025 62 retired married university.degree no yes no
## 32048 73 retired divorced high.school no yes no
## 32056 73 retired divorced high.school no no yes
## 32062 78 retired divorced professional.course no yes no
## 32069 66 retired married professional.course no no no
## 32102 72 retired married professional.course no no no
## 32119 73 retired married unknown no yes no
## 32128 64 retired married university.degree no no no
## 32138 64 retired married university.degree no no no
## 32164 80 retired married basic.4y no no no
## 32175 82 retired married professional.course no yes no
## 32193 66 retired married high.school no unknown unknown
## 32211 75 retired married basic.9y no no no
## 32231 80 retired married basic.4y no yes no
## 32237 78 retired married unknown no no no
## 32261 68 retired divorced basic.4y no yes no
## 32268 64 retired married basic.4y no yes no
## 32289 58 retired married basic.4y no no no
## 32295 71 retired married professional.course no no no
## 32323 74 retired married university.degree no no no
## 32332 68 retired married university.degree no yes no
## 32335 82 retired divorced basic.4y no yes yes
## 32340 82 retired divorced basic.4y no no no
## 32342 74 retired married basic.4y no no no
## 32345 75 retired divorced high.school no no no
## 32350 66 retired divorced basic.4y no yes no
## 32352 80 retired divorced basic.4y no yes no
## 32353 80 retired divorced basic.4y no yes no
## 32354 80 retired divorced basic.4y no yes yes
## 32358 77 retired married high.school no yes no
## 32360 80 retired divorced basic.4y no yes yes
## 32365 73 retired single professional.course no no no
## 32475 77 retired married university.degree no yes no
## 32491 70 retired married basic.4y no yes no
## 32498 68 retired married basic.4y no yes yes
## 32499 76 retired married basic.4y no yes no
## 32510 63 retired married professional.course no yes yes
## 32512 74 retired divorced university.degree no yes no
## 32533 80 retired married basic.4y no no no
## 32538 76 retired divorced basic.4y no yes no
## 32546 69 retired married high.school no no no
## 32547 92 retired divorced unknown unknown no no
## 32551 70 retired divorced high.school no yes no
## 32553 60 retired married high.school no no no
## 32646 89 retired divorced basic.4y no no no
## 32681 78 retired divorced unknown no no no
## 32683 78 retired divorced unknown no no no
## 32718 69 retired divorced professional.course no no no
## 32724 83 retired divorced basic.4y no no yes
## 32749 76 retired married professional.course unknown yes no
## 32750 64 retired married professional.course no no no
## 32767 56 retired married basic.4y no yes no
## 32811 75 retired married university.degree no yes no
## 32826 72 retired married basic.4y no no yes
## 32847 59 retired divorced basic.4y no yes yes
## 32856 73 retired married basic.4y no yes no
## 32858 83 retired married university.degree no yes no
## 32860 83 retired married university.degree no no no
## 32861 66 retired married high.school no yes no
## 32886 78 retired married unknown no yes no
## 32897 80 retired married basic.4y no no no
## 32898 70 retired divorced basic.4y no no no
## 32923 72 retired married basic.4y no yes no
## 32924 78 retired married basic.4y no yes no
## 32927 66 retired married basic.4y no yes no
## 32928 78 retired married basic.4y no yes no
## 32939 62 retired married high.school no no no
## 32947 72 retired married basic.6y no yes no
## 32979 68 retired married university.degree no yes no
## 32980 68 retired married university.degree no no no
## 32983 84 retired divorced basic.4y no yes yes
## 32992 77 retired married basic.4y no unknown unknown
## 32995 68 retired divorced high.school no yes yes
## 33002 51 retired divorced high.school no no no
## 33009 80 retired divorced basic.4y no no yes
## 33028 71 retired married basic.9y no yes yes
## 33040 62 retired married basic.4y no no no
## 33041 67 retired married basic.4y no no no
## 33047 64 retired married professional.course no no no
## 33056 75 retired married unknown no no no
## 33064 61 retired married university.degree no no no
## 33104 59 retired married professional.course no yes no
## 33109 59 retired married professional.course no no no
## 33115 64 retired married basic.4y no unknown unknown
## 33134 92 retired married unknown no no yes
## 33138 65 retired married professional.course no yes no
## 33142 70 retired married basic.4y no yes no
## 33151 92 retired married unknown no no yes
## 33164 76 retired married university.degree no yes no
## 33189 44 retired single high.school no no no
## 33214 76 retired divorced basic.4y no no no
## 33215 60 retired married high.school no yes no
## 33216 56 retired married university.degree no no no
## 33223 69 retired married high.school no yes yes
## 33241 81 retired divorced basic.4y no no no
## 33257 60 retired married university.degree no no no
## 33295 85 retired married basic.4y no no no
## 33296 89 retired divorced basic.4y no yes no
## 33298 66 retired married unknown no yes no
## 33304 86 retired married basic.4y no yes no
## 33333 83 retired divorced basic.4y no no no
## 33335 64 retired divorced basic.4y no yes yes
## 33346 71 retired married high.school no no no
## 33356 84 retired divorced unknown unknown no no
## 33368 60 retired married high.school no no no
## 33372 60 retired married high.school no no no
## 33383 82 retired divorced basic.4y no yes no
## 33390 77 retired married basic.4y no yes no
## 33504 61 retired married basic.4y no no no
## 33511 66 retired married basic.4y no yes no
## 33519 65 retired married basic.4y no yes yes
## 33533 71 retired married professional.course no no no
## 33539 66 retired married basic.4y unknown unknown unknown
## 33540 65 retired married basic.4y no yes yes
## 33560 65 retired married basic.4y no no no
## 33565 68 retired divorced high.school no yes yes
## 33571 68 retired divorced high.school no yes no
## 33583 81 retired married basic.4y no yes no
## 33590 80 retired married professional.course no yes no
## 33608 65 retired married high.school no yes yes
## 33627 65 retired married basic.4y no no no
## 33633 65 retired married basic.4y no yes no
## 33670 63 retired married basic.4y no yes no
## 33733 62 retired married university.degree no yes no
## 33737 62 retired married university.degree no no no
## 33742 73 retired married professional.course no yes no
## contact day_of_week duration campaign pdays previous poutcome
## 3718 telephone tue 1045 1 999 0 nonexistent
## 3875 telephone wed 905 1 999 0 nonexistent
## 4282 telephone fri 924 1 999 0 nonexistent
## 4447 telephone fri 597 6 999 0 nonexistent
## 5561 telephone wed 1730 1 999 0 nonexistent
## 5660 telephone wed 560 9 999 0 nonexistent
## 6689 telephone mon 605 6 999 0 nonexistent
## 7007 telephone wed 633 1 999 0 nonexistent
## 9052 telephone wed 2093 1 999 0 nonexistent
## 9329 telephone thu 1012 1 999 0 nonexistent
## 9685 telephone fri 1094 3 999 0 nonexistent
## 10162 cellular mon 494 4 999 0 nonexistent
## 10220 cellular mon 621 2 999 0 nonexistent
## 11005 cellular wed 1014 1 999 0 nonexistent
## 11383 cellular fri 817 2 999 0 nonexistent
## 11695 cellular mon 1062 4 999 0 nonexistent
## 11861 telephone tue 533 3 999 0 nonexistent
## 12056 cellular wed 767 5 999 0 nonexistent
## 12120 cellular wed 555 1 999 0 nonexistent
## 12656 cellular fri 655 4 999 0 nonexistent
## 14096 cellular fri 1031 8 999 0 nonexistent
## 14130 cellular fri 1448 17 999 0 nonexistent
## 14522 cellular tue 600 1 999 0 nonexistent
## 14609 cellular tue 454 5 999 0 nonexistent
## 15982 cellular thu 674 1 999 0 nonexistent
## 16489 cellular mon 555 1 999 0 nonexistent
## 17004 cellular wed 707 3 999 0 nonexistent
## 17759 cellular tue 933 2 999 0 nonexistent
## 18092 cellular wed 625 5 999 0 nonexistent
## 18696 cellular mon 600 2 999 0 nonexistent
## 18905 cellular tue 388 4 999 0 nonexistent
## 19002 cellular tue 1223 4 999 0 nonexistent
## 19063 cellular tue 556 6 999 0 nonexistent
## 19747 telephone mon 248 1 999 0 nonexistent
## 20190 cellular mon 1132 2 999 0 nonexistent
## 21510 cellular thu 477 1 999 0 nonexistent
## 21776 cellular thu 691 1 999 0 nonexistent
## 22708 cellular mon 187 3 999 0 nonexistent
## 22783 cellular wed 796 5 999 0 nonexistent
## 22787 cellular wed 126 1 999 0 nonexistent
## 22789 cellular wed 101 7 999 0 nonexistent
## 22792 cellular wed 188 3 999 0 nonexistent
## 22793 cellular wed 101 1 999 0 nonexistent
## 22798 cellular thu 156 1 999 0 nonexistent
## 22831 cellular wed 201 1 999 0 nonexistent
## 22882 cellular fri 179 1 999 0 nonexistent
## 22931 cellular wed 387 2 999 0 nonexistent
## 23041 cellular mon 158 1 999 0 nonexistent
## 23078 cellular tue 465 4 999 0 nonexistent
## 23173 cellular wed 150 1 999 1 failure
## 23174 telephone wed 236 1 999 0 nonexistent
## 23189 cellular wed 321 1 999 0 nonexistent
## 23208 cellular wed 624 1 999 0 nonexistent
## 23304 cellular tue 342 1 999 0 nonexistent
## 23323 cellular tue 1073 1 999 0 nonexistent
## 23350 cellular tue 216 1 999 0 nonexistent
## 23355 cellular tue 207 2 999 0 nonexistent
## 23362 cellular wed 223 2 999 0 nonexistent
## 23370 cellular wed 167 2 999 1 failure
## 23376 cellular wed 282 2 999 0 nonexistent
## 23404 cellular wed 512 11 999 0 nonexistent
## 23409 cellular wed 140 1 999 0 nonexistent
## 23420 cellular wed 129 1 999 0 nonexistent
## 23443 cellular thu 144 4 999 0 nonexistent
## 23530 cellular thu 346 2 999 0 nonexistent
## 23919 cellular fri 188 3 999 0 nonexistent
## 24043 cellular fri 314 2 999 0 nonexistent
## 24204 cellular mon 194 1 999 0 nonexistent
## 24293 cellular mon 349 1 999 0 nonexistent
## 24306 cellular mon 227 4 999 0 nonexistent
## 24424 cellular mon 453 1 999 0 nonexistent
## 24538 cellular tue 146 4 999 0 nonexistent
## 24544 cellular tue 274 1 999 0 nonexistent
## 24545 telephone tue 1348 4 999 0 nonexistent
## 24546 cellular tue 109 1 999 1 failure
## 24550 cellular tue 164 1 999 0 nonexistent
## 24558 cellular tue 106 1 999 0 nonexistent
## 24564 cellular tue 356 2 999 1 failure
## 24568 telephone tue 137 2 999 0 nonexistent
## 24570 cellular tue 129 3 999 0 nonexistent
## 24571 cellular tue 157 4 999 0 nonexistent
## 24589 cellular wed 104 1 999 2 failure
## 24599 cellular wed 245 4 999 1 failure
## 24600 cellular wed 124 2 999 0 nonexistent
## 24606 cellular wed 1288 3 999 0 nonexistent
## 24675 cellular thu 510 1 999 0 nonexistent
## 24677 cellular thu 968 1 5 2 failure
## 24740 cellular thu 381 1 999 1 failure
## 24809 cellular thu 616 1 999 0 nonexistent
## 24824 cellular thu 430 1 999 0 nonexistent
## 24867 cellular thu 266 1 999 1 failure
## 24903 cellular thu 949 2 999 0 nonexistent
## 24980 cellular mon 228 6 999 0 nonexistent
## 25111 cellular tue 762 3 999 0 nonexistent
## 29350 cellular fri 176 1 999 0 nonexistent
## 29489 cellular tue 133 1 3 1 success
## 29630 cellular wed 108 2 999 0 nonexistent
## 29705 cellular mon 1460 1 999 0 nonexistent
## 29710 cellular mon 579 1 999 0 nonexistent
## 29764 cellular tue 663 1 999 0 nonexistent
## 29882 telephone wed 437 1 999 0 nonexistent
## 30093 cellular mon 107 3 999 0 nonexistent
## 30122 cellular fri 186 2 999 0 nonexistent
## 30155 cellular mon 257 1 999 0 nonexistent
## 30286 cellular fri 247 2 999 0 nonexistent
## 30307 cellular fri 308 2 999 0 nonexistent
## 30405 cellular tue 219 1 999 0 nonexistent
## 30437 cellular wed 385 2 999 1 failure
## 30443 cellular thu 155 2 4 1 success
## 30452 cellular thu 427 1 15 1 success
## 30461 cellular thu 1452 1 999 1 failure
## 30473 cellular thu 550 1 999 0 nonexistent
## 30474 cellular fri 345 2 999 0 nonexistent
## 30485 cellular fri 714 1 999 0 nonexistent
## 30493 cellular fri 530 2 999 0 nonexistent
## 30498 cellular fri 453 1 999 0 nonexistent
## 30522 cellular fri 233 2 999 0 nonexistent
## 30588 cellular tue 140 1 999 0 nonexistent
## 30601 cellular tue 125 1 6 1 success
## 30637 cellular wed 239 2 999 1 failure
## 30640 cellular wed 147 2 999 1 failure
## 30724 cellular thu 204 3 999 1 failure
## 30755 cellular fri 172 4 999 0 nonexistent
## 30776 cellular fri 137 1 3 1 success
## 30778 cellular fri 125 1 999 0 nonexistent
## 30780 cellular fri 94 4 999 0 nonexistent
## 30804 cellular mon 348 1 999 0 nonexistent
## 30805 cellular mon 186 1 999 0 nonexistent
## 30806 telephone mon 199 1 999 0 nonexistent
## 30827 cellular mon 156 1 999 1 failure
## 30829 cellular tue 460 2 999 0 nonexistent
## 30833 telephone tue 249 2 3 1 success
## 30850 cellular tue 216 1 999 0 nonexistent
## 30854 cellular tue 301 1 999 1 failure
## 30859 cellular wed 536 1 13 1 success
## 30872 cellular wed 232 3 999 0 nonexistent
## 30875 cellular wed 145 1 3 2 success
## 30886 cellular thu 267 1 999 0 nonexistent
## 30896 cellular thu 217 1 6 2 failure
## 30912 cellular fri 374 1 999 0 nonexistent
## 30913 cellular fri 168 1 15 1 success
## 30916 cellular fri 273 1 999 0 nonexistent
## 30932 cellular fri 252 4 999 0 nonexistent
## 30963 cellular tue 130 2 999 0 nonexistent
## 30966 cellular tue 95 2 999 0 nonexistent
## 30968 cellular tue 531 3 6 1 success
## 30970 cellular tue 242 6 3 1 success
## 30978 telephone wed 128 3 999 0 nonexistent
## 30993 cellular thu 398 2 999 1 failure
## 31010 cellular fri 517 2 999 1 failure
## 31023 cellular mon 160 3 999 1 failure
## 31053 cellular tue 181 1 999 1 failure
## 31107 cellular mon 350 1 3 2 success
## 31131 cellular tue 206 1 999 0 nonexistent
## 31137 cellular tue 191 1 999 1 failure
## 31146 cellular wed 394 1 999 0 nonexistent
## 31160 cellular wed 259 2 3 1 success
## 31184 cellular fri 658 4 999 0 nonexistent
## 31193 cellular tue 150 1 3 2 success
## 31237 cellular fri 692 1 999 0 nonexistent
## 31246 cellular fri 210 1 999 0 nonexistent
## 31254 cellular wed 117 1 3 1 success
## 31275 cellular fri 849 2 4 1 success
## 31279 cellular tue 215 1 999 0 nonexistent
## 31282 cellular tue 144 1 999 0 nonexistent
## 31286 cellular tue 384 2 999 0 nonexistent
## 31306 cellular wed 335 1 999 0 nonexistent
## 31326 cellular mon 348 1 999 0 nonexistent
## 31340 cellular tue 242 1 999 2 failure
## 31363 cellular wed 251 1 999 0 nonexistent
## 31387 cellular thu 134 2 6 1 success
## 31419 cellular fri 717 2 999 0 nonexistent
## 31423 cellular mon 131 1 999 0 nonexistent
## 31427 cellular mon 98 1 999 0 nonexistent
## 31479 cellular thu 133 1 3 1 success
## 31495 cellular fri 476 1 2 2 success
## 31498 cellular fri 272 2 999 0 nonexistent
## 31525 cellular tue 102 1 6 2 success
## 31544 cellular tue 353 1 999 0 nonexistent
## 31551 cellular wed 96 1 999 0 nonexistent
## 31567 cellular wed 207 1 6 1 success
## 31573 cellular thu 139 1 999 0 nonexistent
## 31574 cellular thu 169 1 999 0 nonexistent
## 31581 cellular thu 189 3 999 0 nonexistent
## 31582 cellular thu 251 2 999 0 nonexistent
## 31594 cellular thu 124 1 999 0 nonexistent
## 31598 cellular thu 377 1 999 0 nonexistent
## 31615 cellular fri 257 3 999 0 nonexistent
## 31652 cellular tue 102 1 999 0 nonexistent
## 31660 cellular tue 149 1 999 0 nonexistent
## 31671 telephone tue 383 1 999 0 nonexistent
## 31684 cellular thu 749 7 999 0 nonexistent
## 31686 cellular thu 355 3 999 0 nonexistent
## 31695 cellular mon 178 1 999 0 nonexistent
## 31703 cellular mon 347 4 6 1 success
## 31718 cellular tue 301 2 3 1 success
## 31721 cellular tue 594 1 3 1 success
## 31734 cellular wed 406 1 999 1 failure
## 31738 cellular wed 199 1 999 0 nonexistent
## 31756 cellular thu 253 1 999 2 failure
## 31763 cellular thu 99 1 999 0 nonexistent
## 31774 telephone thu 1127 1 999 0 nonexistent
## 31790 cellular fri 186 2 3 1 success
## 31793 cellular fri 226 1 3 3 success
## 31800 cellular fri 193 1 3 1 success
## 31805 cellular fri 182 2 3 1 success
## 31811 cellular mon 301 1 9 3 failure
## 31814 cellular mon 146 1 3 1 success
## 31815 telephone mon 167 1 999 0 nonexistent
## 31840 cellular mon 257 1 999 0 nonexistent
## 31842 cellular mon 445 2 999 1 failure
## 31847 cellular mon 1248 2 999 1 failure
## 31853 cellular tue 134 2 3 1 success
## 31859 cellular tue 525 3 999 0 nonexistent
## 31861 cellular tue 242 1 3 3 success
## 31872 cellular wed 968 2 3 3 success
## 31888 cellular wed 234 1 999 0 nonexistent
## 31889 cellular wed 487 1 999 2 failure
## 31890 cellular wed 232 1 3 1 success
## 31901 cellular fri 319 2 999 1 failure
## 31905 cellular fri 160 1 999 0 nonexistent
## 31923 cellular thu 155 1 4 3 success
## 31926 cellular thu 472 1 999 0 nonexistent
## 31931 cellular fri 439 3 6 2 success
## 31964 telephone wed 343 2 999 0 nonexistent
## 31972 cellular thu 429 1 999 0 nonexistent
## 31980 cellular fri 166 3 999 0 nonexistent
## 31989 cellular mon 321 3 6 1 success
## 31996 cellular wed 354 1 999 0 nonexistent
## 31997 cellular wed 546 2 999 0 nonexistent
## 32003 cellular thu 330 2 4 2 success
## 32005 cellular thu 324 3 4 2 success
## 32025 cellular wed 282 1 6 1 success
## 32048 cellular tue 284 1 999 0 nonexistent
## 32056 cellular tue 63 3 6 2 success
## 32062 cellular tue 591 1 999 1 failure
## 32069 cellular tue 177 1 999 1 failure
## 32102 cellular fri 87 1 3 1 success
## 32119 telephone mon 659 1 999 0 nonexistent
## 32128 cellular tue 139 2 6 1 success
## 32138 cellular tue 700 2 5 1 success
## 32164 cellular fri 213 3 6 4 success
## 32175 telephone mon 506 2 999 0 nonexistent
## 32193 cellular tue 881 3 999 1 failure
## 32211 cellular mon 293 2 999 1 failure
## 32231 cellular thu 156 1 999 0 nonexistent
## 32237 cellular thu 272 1 6 2 success
## 32261 cellular tue 277 2 11 1 success
## 32268 cellular wed 262 2 999 0 nonexistent
## 32289 cellular wed 1307 1 999 0 nonexistent
## 32295 cellular thu 559 1 999 0 nonexistent
## 32323 cellular thu 200 1 999 0 nonexistent
## 32332 cellular fri 330 3 999 0 nonexistent
## 32335 cellular mon 125 2 999 0 nonexistent
## 32340 cellular mon 529 1 6 2 success
## 32342 telephone mon 1143 5 999 0 nonexistent
## 32345 telephone tue 162 1 6 2 success
## 32350 cellular wed 476 1 999 0 nonexistent
## 32352 telephone wed 403 1 999 0 nonexistent
## 32353 telephone wed 623 2 999 0 nonexistent
## 32354 cellular wed 654 2 999 0 nonexistent
## 32358 cellular thu 165 7 999 0 nonexistent
## 32360 cellular thu 554 1 10 2 success
## 32365 cellular fri 291 1 6 3 success
## 32475 cellular wed 152 1 6 4 success
## 32491 cellular thu 331 1 3 2 success
## 32498 cellular fri 220 1 6 1 success
## 32499 cellular fri 1205 2 6 2 success
## 32510 cellular mon 444 2 14 1 success
## 32512 cellular tue 387 1 999 0 nonexistent
## 32533 cellular mon 382 1 3 3 success
## 32538 cellular mon 168 1 999 1 failure
## 32546 cellular wed 289 1 10 1 success
## 32547 cellular wed 405 3 999 1 failure
## 32551 telephone wed 283 2 4 2 success
## 32553 cellular thu 181 3 6 1 success
## 32646 cellular mon 245 1 999 0 nonexistent
## 32681 cellular thu 282 4 999 0 nonexistent
## 32683 cellular thu 544 1 999 0 nonexistent
## 32718 cellular fri 213 6 12 2 failure
## 32724 cellular fri 472 2 999 0 nonexistent
## 32749 cellular tue 352 1 3 1 success
## 32750 cellular tue 222 1 999 3 failure
## 32767 cellular wed 337 2 3 2 success
## 32811 cellular thu 229 1 999 2 failure
## 32826 cellular fri 483 4 8 1 success
## 32847 cellular mon 796 1 6 1 success
## 32856 cellular tue 305 1 999 0 nonexistent
## 32858 cellular tue 178 1 6 2 success
## 32860 telephone tue 617 1 12 1 success
## 32861 cellular tue 317 1 999 0 nonexistent
## 32886 cellular wed 87 3 999 0 nonexistent
## 32897 telephone thu 552 2 999 0 nonexistent
## 32898 cellular thu 390 2 6 2 success
## 32923 cellular mon 268 1 999 0 nonexistent
## 32924 cellular mon 173 2 999 1 failure
## 32927 cellular mon 568 7 999 0 nonexistent
## 32928 cellular mon 212 1 13 2 success
## 32939 cellular tue 355 2 999 0 nonexistent
## 32947 cellular wed 143 1 999 0 nonexistent
## 32979 cellular mon 546 2 999 1 failure
## 32980 cellular mon 414 2 12 1 success
## 32983 cellular tue 666 1 3 2 success
## 32992 cellular tue 218 2 3 1 success
## 32995 cellular wed 340 1 3 1 success
## 33002 cellular wed 115 3 999 0 nonexistent
## 33009 cellular thu 169 2 6 2 success
## 33028 cellular fri 230 6 999 0 nonexistent
## 33040 cellular mon 317 1 9 2 failure
## 33041 cellular mon 341 2 13 1 success
## 33047 cellular tue 482 1 999 0 nonexistent
## 33056 telephone tue 676 2 999 4 failure
## 33064 telephone wed 250 2 7 1 success
## 33104 cellular fri 303 2 3 2 success
## 33109 cellular fri 218 3 3 1 success
## 33115 telephone fri 245 3 999 0 nonexistent
## 33134 cellular tue 1064 1 3 1 success
## 33138 cellular tue 261 1 6 3 success
## 33142 cellular tue 356 3 6 1 success
## 33151 cellular wed 370 1 3 4 success
## 33164 cellular thu 504 2 6 3 success
## 33189 cellular mon 2035 4 999 0 nonexistent
## 33214 cellular tue 185 1 999 1 failure
## 33215 cellular tue 443 2 999 2 failure
## 33216 cellular tue 634 3 999 1 failure
## 33223 cellular wed 840 1 6 2 success
## 33241 cellular wed 532 2 7 1 success
## 33257 cellular thu 799 2 999 0 nonexistent
## 33295 cellular tue 728 1 3 2 success
## 33296 cellular tue 314 1 999 0 nonexistent
## 33298 cellular tue 1394 2 6 1 success
## 33304 cellular tue 288 3 999 0 nonexistent
## 33333 cellular thu 268 1 9 3 success
## 33335 telephone fri 211 2 6 2 success
## 33346 cellular mon 363 1 999 0 nonexistent
## 33356 cellular tue 333 1 3 2 success
## 33368 cellular wed 1640 1 999 0 nonexistent
## 33372 cellular wed 200 1 6 1 success
## 33383 cellular thu 143 2 999 1 failure
## 33390 cellular thu 190 1 999 0 nonexistent
## 33504 telephone fri 194 1 999 0 nonexistent
## 33511 telephone fri 369 2 999 1 failure
## 33519 cellular mon 165 1 3 1 success
## 33533 cellular tue 323 1 999 0 nonexistent
## 33539 cellular wed 216 1 6 3 success
## 33540 cellular wed 190 1 3 3 success
## 33560 cellular fri 187 2 7 3 success
## 33565 cellular mon 567 1 3 1 success
## 33571 cellular mon 222 1 999 0 nonexistent
## 33583 cellular wed 621 1 999 2 failure
## 33590 cellular thu 411 1 999 1 failure
## 33608 cellular fri 344 2 12 1 success
## 33627 cellular wed 882 1 999 0 nonexistent
## 33633 cellular wed 258 1 3 3 success
## 33670 cellular fri 413 1 3 2 success
## 33733 cellular thu 329 1 999 2 failure
## 33737 cellular thu 483 2 6 3 success
## 33742 cellular fri 334 1 999 0 nonexistent
## emp.var.rate cons.price.idx cons.conf.idx euribor3m nr.employed y
## 3718 1.1 93.994 -36.4 4.856 5191.0 yes
## 3875 1.1 93.994 -36.4 4.858 5191.0 yes
## 4282 1.1 93.994 -36.4 4.857 5191.0 yes
## 4447 1.1 93.994 -36.4 4.857 5191.0 yes
## 5561 1.1 93.994 -36.4 4.857 5191.0 yes
## 5660 1.1 93.994 -36.4 4.857 5191.0 yes
## 6689 1.4 94.465 -41.8 4.865 5228.1 yes
## 7007 1.4 94.465 -41.8 4.864 5228.1 yes
## 9052 1.4 94.465 -41.8 4.962 5228.1 yes
## 9329 1.4 94.465 -41.8 4.961 5228.1 yes
## 9685 1.4 94.465 -41.8 4.959 5228.1 yes
## 10162 1.4 93.918 -42.7 4.960 5228.1 yes
## 10220 1.4 93.918 -42.7 4.960 5228.1 yes
## 11005 1.4 93.918 -42.7 4.962 5228.1 yes
## 11383 1.4 93.918 -42.7 4.963 5228.1 yes
## 11695 1.4 93.918 -42.7 4.962 5228.1 yes
## 11861 1.4 93.918 -42.7 4.961 5228.1 yes
## 12056 1.4 93.918 -42.7 4.957 5228.1 yes
## 12120 1.4 93.918 -42.7 4.957 5228.1 yes
## 12656 1.4 93.918 -42.7 4.957 5228.1 yes
## 14096 1.4 93.918 -42.7 4.962 5228.1 yes
## 14130 1.4 93.918 -42.7 4.962 5228.1 yes
## 14522 1.4 93.918 -42.7 4.961 5228.1 yes
## 14609 1.4 93.918 -42.7 4.961 5228.1 yes
## 15982 1.4 93.444 -36.1 4.968 5228.1 yes
## 16489 1.4 93.444 -36.1 4.965 5228.1 yes
## 17004 1.4 93.444 -36.1 4.965 5228.1 yes
## 17759 1.4 93.444 -36.1 4.963 5228.1 yes
## 18092 1.4 93.444 -36.1 4.964 5228.1 yes
## 18696 1.4 93.444 -36.1 4.965 5228.1 yes
## 18905 1.4 93.444 -36.1 4.965 5228.1 yes
## 19002 1.4 93.444 -36.1 4.965 5228.1 yes
## 19063 1.4 93.444 -36.1 4.965 5228.1 yes
## 19747 -0.1 93.200 -42.0 4.406 5195.8 yes
## 20190 -0.1 93.200 -42.0 4.191 5195.8 yes
## 21510 -0.1 93.200 -42.0 4.076 5195.8 yes
## 21776 -0.1 93.200 -42.0 4.076 5195.8 yes
## 22708 -1.8 92.843 -50.0 1.811 5099.1 yes
## 22783 -1.8 92.843 -50.0 1.663 5099.1 yes
## 22787 -1.8 92.843 -50.0 1.663 5099.1 yes
## 22789 -1.8 92.843 -50.0 1.663 5099.1 yes
## 22792 -1.8 92.843 -50.0 1.663 5099.1 yes
## 22793 -1.8 92.843 -50.0 1.663 5099.1 yes
## 22798 -1.8 92.843 -50.0 1.650 5099.1 yes
## 22831 -1.8 92.843 -50.0 1.602 5099.1 yes
## 22882 -1.8 92.843 -50.0 1.531 5099.1 yes
## 22931 -1.8 93.075 -47.1 1.498 5099.1 yes
## 23041 -1.8 93.075 -47.1 1.466 5099.1 yes
## 23078 -1.8 93.075 -47.1 1.453 5099.1 yes
## 23173 -1.8 93.075 -47.1 1.445 5099.1 yes
## 23174 -1.8 93.075 -47.1 1.445 5099.1 yes
## 23189 -1.8 93.075 -47.1 1.445 5099.1 yes
## 23208 -1.8 93.075 -47.1 1.445 5099.1 yes
## 23304 -1.8 93.075 -47.1 1.423 5099.1 yes
## 23323 -1.8 93.075 -47.1 1.423 5099.1 yes
## 23350 -1.8 93.075 -47.1 1.423 5099.1 yes
## 23355 -1.8 93.075 -47.1 1.423 5099.1 yes
## 23362 -1.8 93.075 -47.1 1.415 5099.1 yes
## 23370 -1.8 93.075 -47.1 1.415 5099.1 yes
## 23376 -1.8 93.075 -47.1 1.415 5099.1 yes
## 23404 -1.8 93.075 -47.1 1.415 5099.1 yes
## 23409 -1.8 93.075 -47.1 1.415 5099.1 yes
## 23420 -1.8 93.075 -47.1 1.415 5099.1 yes
## 23443 -1.8 93.075 -47.1 1.410 5099.1 yes
## 23530 -1.8 93.075 -47.1 1.410 5099.1 yes
## 23919 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24043 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24204 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24293 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24306 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24424 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24538 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24544 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24545 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24546 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24550 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24558 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24564 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24568 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24570 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24571 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24589 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24599 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24600 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24606 -1.8 93.075 -47.1 1.405 5099.1 yes
## 24675 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24677 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24740 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24809 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24824 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24867 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24903 -1.8 93.075 -47.1 1.365 5099.1 yes
## 24980 -1.8 92.893 -46.2 1.354 5099.1 yes
## 25111 -1.8 92.893 -46.2 1.344 5099.1 yes
## 29350 -1.8 92.893 -46.2 1.259 5099.1 yes
## 29489 -1.8 92.893 -46.2 1.266 5099.1 yes
## 29630 -1.8 92.893 -46.2 1.270 5099.1 yes
## 29705 -2.9 92.963 -40.8 1.266 5076.2 yes
## 29710 -2.9 92.963 -40.8 1.266 5076.2 yes
## 29764 -2.9 92.963 -40.8 1.262 5076.2 yes
## 29882 -2.9 92.963 -40.8 1.260 5076.2 yes
## 30093 -2.9 92.963 -40.8 1.281 5076.2 yes
## 30122 -2.9 92.963 -40.8 1.268 5076.2 yes
## 30155 -2.9 92.963 -40.8 1.260 5076.2 yes
## 30286 -2.9 92.469 -33.6 1.059 5076.2 yes
## 30307 -2.9 92.469 -33.6 1.059 5076.2 yes
## 30405 -2.9 92.201 -31.4 0.884 5076.2 yes
## 30437 -2.9 92.201 -31.4 0.884 5076.2 yes
## 30443 -2.9 92.201 -31.4 0.883 5076.2 yes
## 30452 -2.9 92.201 -31.4 0.883 5076.2 yes
## 30461 -2.9 92.201 -31.4 0.883 5076.2 yes
## 30473 -2.9 92.201 -31.4 0.883 5076.2 yes
## 30474 -2.9 92.201 -31.4 0.881 5076.2 yes
## 30485 -2.9 92.201 -31.4 0.881 5076.2 yes
## 30493 -2.9 92.201 -31.4 0.881 5076.2 yes
## 30498 -2.9 92.201 -31.4 0.881 5076.2 yes
## 30522 -2.9 92.201 -31.4 0.881 5076.2 yes
## 30588 -2.9 92.201 -31.4 0.883 5076.2 yes
## 30601 -2.9 92.201 -31.4 0.883 5076.2 yes
## 30637 -2.9 92.201 -31.4 0.879 5076.2 yes
## 30640 -2.9 92.201 -31.4 0.879 5076.2 yes
## 30724 -2.9 92.201 -31.4 0.873 5076.2 yes
## 30755 -2.9 92.201 -31.4 0.869 5076.2 yes
## 30776 -2.9 92.201 -31.4 0.869 5076.2 yes
## 30778 -2.9 92.201 -31.4 0.869 5076.2 yes
## 30780 -2.9 92.201 -31.4 0.869 5076.2 yes
## 30804 -2.9 92.201 -31.4 0.861 5076.2 yes
## 30805 -2.9 92.201 -31.4 0.861 5076.2 yes
## 30806 -2.9 92.201 -31.4 0.861 5076.2 yes
## 30827 -2.9 92.201 -31.4 0.861 5076.2 yes
## 30829 -2.9 92.201 -31.4 0.859 5076.2 yes
## 30833 -2.9 92.201 -31.4 0.859 5076.2 yes
## 30850 -2.9 92.201 -31.4 0.859 5076.2 yes
## 30854 -2.9 92.201 -31.4 0.859 5076.2 yes
## 30859 -2.9 92.201 -31.4 0.854 5076.2 yes
## 30872 -2.9 92.201 -31.4 0.854 5076.2 yes
## 30875 -2.9 92.201 -31.4 0.854 5076.2 yes
## 30886 -2.9 92.201 -31.4 0.851 5076.2 yes
## 30896 -2.9 92.201 -31.4 0.851 5076.2 yes
## 30912 -2.9 92.201 -31.4 0.849 5076.2 yes
## 30913 -2.9 92.201 -31.4 0.849 5076.2 yes
## 30916 -2.9 92.201 -31.4 0.849 5076.2 yes
## 30932 -2.9 92.201 -31.4 0.849 5076.2 yes
## 30963 -2.9 92.201 -31.4 0.838 5076.2 yes
## 30966 -2.9 92.201 -31.4 0.838 5076.2 yes
## 30968 -2.9 92.201 -31.4 0.838 5076.2 yes
## 30970 -2.9 92.201 -31.4 0.838 5076.2 yes
## 30978 -2.9 92.201 -31.4 0.834 5076.2 yes
## 30993 -2.9 92.201 -31.4 0.829 5076.2 yes
## 31010 -2.9 92.201 -31.4 0.825 5076.2 yes
## 31023 -2.9 92.201 -31.4 0.821 5076.2 yes
## 31053 -3.4 92.379 -29.8 0.819 5017.5 yes
## 31107 -3.4 92.379 -29.8 0.797 5017.5 yes
## 31131 -3.4 92.379 -29.8 0.788 5017.5 yes
## 31137 -3.4 92.379 -29.8 0.788 5017.5 yes
## 31146 -3.4 92.379 -29.8 0.781 5017.5 yes
## 31160 -3.4 92.379 -29.8 0.781 5017.5 yes
## 31184 -3.4 92.379 -29.8 0.773 5017.5 yes
## 31193 -3.4 92.379 -29.8 0.770 5017.5 yes
## 31237 -3.4 92.379 -29.8 0.741 5017.5 yes
## 31246 -3.4 92.379 -29.8 0.741 5017.5 yes
## 31254 -3.4 92.379 -29.8 0.753 5017.5 yes
## 31275 -3.4 92.431 -26.9 0.752 5017.5 yes
## 31279 -3.4 92.431 -26.9 0.744 5017.5 yes
## 31282 -3.4 92.431 -26.9 0.744 5017.5 yes
## 31286 -3.4 92.431 -26.9 0.744 5017.5 yes
## 31306 -3.4 92.431 -26.9 0.740 5017.5 yes
## 31326 -3.4 92.431 -26.9 0.743 5017.5 yes
## 31340 -3.4 92.431 -26.9 0.742 5017.5 yes
## 31363 -3.4 92.431 -26.9 0.742 5017.5 yes
## 31387 -3.4 92.431 -26.9 0.740 5017.5 yes
## 31419 -3.4 92.431 -26.9 0.739 5017.5 yes
## 31423 -3.4 92.431 -26.9 0.739 5017.5 yes
## 31427 -3.4 92.431 -26.9 0.739 5017.5 yes
## 31479 -3.4 92.431 -26.9 0.733 5017.5 yes
## 31495 -3.4 92.431 -26.9 0.730 5017.5 yes
## 31498 -3.4 92.431 -26.9 0.730 5017.5 yes
## 31525 -3.4 92.431 -26.9 0.728 5017.5 yes
## 31544 -3.4 92.431 -26.9 0.728 5017.5 yes
## 31551 -3.4 92.431 -26.9 0.724 5017.5 yes
## 31567 -3.4 92.431 -26.9 0.724 5017.5 yes
## 31573 -3.4 92.431 -26.9 0.722 5017.5 yes
## 31574 -3.4 92.431 -26.9 0.722 5017.5 yes
## 31581 -3.4 92.431 -26.9 0.722 5017.5 yes
## 31582 -3.4 92.431 -26.9 0.722 5017.5 yes
## 31594 -3.4 92.431 -26.9 0.722 5017.5 yes
## 31598 -3.4 92.431 -26.9 0.722 5017.5 yes
## 31615 -3.4 92.431 -26.9 0.720 5017.5 yes
## 31652 -3.4 92.649 -30.1 0.720 5017.5 yes
## 31660 -3.4 92.649 -30.1 0.720 5017.5 yes
## 31671 -3.4 92.649 -30.1 0.720 5017.5 yes
## 31684 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31686 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31695 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31703 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31718 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31721 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31734 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31738 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31756 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31763 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31774 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31790 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31793 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31800 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31805 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31811 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31814 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31815 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31840 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31842 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31847 -3.4 92.649 -30.1 0.714 5017.5 yes
## 31853 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31859 -3.4 92.649 -30.1 0.715 5017.5 yes
## 31861 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31872 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31888 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31889 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31890 -3.4 92.649 -30.1 0.716 5017.5 yes
## 31901 -3.4 92.649 -30.1 0.718 5017.5 yes
## 31905 -3.4 92.649 -30.1 0.718 5017.5 yes
## 31923 -3.0 92.713 -33.0 0.720 5023.5 yes
## 31926 -3.0 92.713 -33.0 0.720 5023.5 yes
## 31931 -3.0 92.713 -33.0 0.718 5023.5 yes
## 31964 -3.0 92.713 -33.0 0.715 5023.5 yes
## 31972 -3.0 92.713 -33.0 0.714 5023.5 yes
## 31980 -3.0 92.713 -33.0 0.714 5023.5 yes
## 31989 -3.0 92.713 -33.0 0.715 5023.5 yes
## 31996 -3.0 92.713 -33.0 0.715 5023.5 yes
## 31997 -3.0 92.713 -33.0 0.715 5023.5 yes
## 32003 -3.0 92.713 -33.0 0.712 5023.5 yes
## 32005 -3.0 92.713 -33.0 0.712 5023.5 yes
## 32025 -3.0 92.713 -33.0 0.706 5023.5 yes
## 32048 -3.0 92.713 -33.0 0.707 5023.5 yes
## 32056 -3.0 92.713 -33.0 0.707 5023.5 yes
## 32062 -1.8 93.369 -34.8 0.655 5008.7 yes
## 32069 -1.8 93.369 -34.8 0.655 5008.7 yes
## 32102 -1.8 93.369 -34.8 0.653 5008.7 yes
## 32119 -1.8 93.369 -34.8 0.652 5008.7 yes
## 32128 -1.8 93.369 -34.8 0.652 5008.7 yes
## 32138 -1.8 93.369 -34.8 0.652 5008.7 yes
## 32164 -1.8 93.369 -34.8 0.649 5008.7 yes
## 32175 -1.8 93.369 -34.8 0.646 5008.7 yes
## 32193 -1.8 93.369 -34.8 0.646 5008.7 yes
## 32211 -1.8 93.369 -34.8 0.639 5008.7 yes
## 32231 -1.8 93.369 -34.8 0.635 5008.7 yes
## 32237 -1.8 93.369 -34.8 0.635 5008.7 yes
## 32261 -1.8 93.369 -34.8 0.635 5008.7 yes
## 32268 -1.8 93.369 -34.8 0.634 5008.7 yes
## 32289 -1.8 93.749 -34.6 0.639 5008.7 yes
## 32295 -1.8 93.749 -34.6 0.640 5008.7 yes
## 32323 -1.8 93.749 -34.6 0.644 5008.7 yes
## 32332 -1.8 93.749 -34.6 0.644 5008.7 yes
## 32335 -1.8 93.749 -34.6 0.643 5008.7 yes
## 32340 -1.8 93.749 -34.6 0.643 5008.7 yes
## 32342 -1.8 93.749 -34.6 0.643 5008.7 yes
## 32345 -1.8 93.749 -34.6 0.642 5008.7 yes
## 32350 -1.8 93.749 -34.6 0.642 5008.7 yes
## 32352 -1.8 93.749 -34.6 0.642 5008.7 yes
## 32353 -1.8 93.749 -34.6 0.642 5008.7 yes
## 32354 -1.8 93.749 -34.6 0.642 5008.7 yes
## 32358 -1.8 93.749 -34.6 0.644 5008.7 yes
## 32360 -1.8 93.749 -34.6 0.644 5008.7 yes
## 32365 -1.8 93.749 -34.6 0.645 5008.7 yes
## 32475 -1.8 93.876 -40.0 0.682 5008.7 yes
## 32491 -1.8 93.876 -40.0 0.683 5008.7 yes
## 32498 -1.8 93.876 -40.0 0.684 5008.7 yes
## 32499 -1.8 93.876 -40.0 0.684 5008.7 yes
## 32510 -1.8 93.876 -40.0 0.685 5008.7 yes
## 32512 -1.8 93.876 -40.0 0.688 5008.7 yes
## 32533 -1.8 93.876 -40.0 0.697 5008.7 yes
## 32538 -1.8 93.876 -40.0 0.697 5008.7 yes
## 32546 -1.8 93.876 -40.0 0.697 5008.7 yes
## 32547 -1.8 93.876 -40.0 0.697 5008.7 yes
## 32551 -1.8 93.876 -40.0 0.697 5008.7 yes
## 32553 -1.8 93.876 -40.0 0.699 5008.7 yes
## 32646 -1.7 94.055 -39.8 0.720 4991.6 yes
## 32681 -1.7 94.055 -39.8 0.729 4991.6 yes
## 32683 -1.7 94.055 -39.8 0.729 4991.6 yes
## 32718 -1.7 94.055 -39.8 0.748 4991.6 yes
## 32724 -1.7 94.055 -39.8 0.748 4991.6 yes
## 32749 -1.7 94.055 -39.8 0.761 4991.6 yes
## 32750 -1.7 94.055 -39.8 0.761 4991.6 yes
## 32767 -1.7 94.055 -39.8 0.767 4991.6 yes
## 32811 -1.7 94.215 -40.3 0.810 4991.6 yes
## 32826 -1.7 94.215 -40.3 0.822 4991.6 yes
## 32847 -1.7 94.215 -40.3 0.827 4991.6 yes
## 32856 -1.7 94.215 -40.3 0.835 4991.6 yes
## 32858 -1.7 94.215 -40.3 0.835 4991.6 yes
## 32860 -1.7 94.215 -40.3 0.835 4991.6 yes
## 32861 -1.7 94.215 -40.3 0.835 4991.6 yes
## 32886 -1.7 94.215 -40.3 0.840 4991.6 yes
## 32897 -1.7 94.215 -40.3 0.846 4991.6 yes
## 32898 -1.7 94.215 -40.3 0.846 4991.6 yes
## 32923 -1.7 94.215 -40.3 0.870 4991.6 yes
## 32924 -1.7 94.215 -40.3 0.870 4991.6 yes
## 32927 -1.7 94.215 -40.3 0.870 4991.6 yes
## 32928 -1.7 94.215 -40.3 0.870 4991.6 yes
## 32939 -1.7 94.215 -40.3 0.876 4991.6 yes
## 32947 -1.7 94.215 -40.3 0.881 4991.6 yes
## 32979 -1.7 94.215 -40.3 0.889 4991.6 yes
## 32980 -1.7 94.215 -40.3 0.889 4991.6 yes
## 32983 -1.7 94.215 -40.3 0.893 4991.6 yes
## 32992 -1.7 94.215 -40.3 0.893 4991.6 yes
## 32995 -1.7 94.215 -40.3 0.896 4991.6 yes
## 33002 -1.7 94.215 -40.3 0.896 4991.6 yes
## 33009 -1.7 94.215 -40.3 0.899 4991.6 yes
## 33028 -1.7 94.215 -40.3 0.896 4991.6 yes
## 33040 -1.7 94.027 -38.3 0.898 4991.6 yes
## 33041 -1.7 94.027 -38.3 0.898 4991.6 yes
## 33047 -1.7 94.027 -38.3 0.899 4991.6 yes
## 33056 -1.7 94.027 -38.3 0.899 4991.6 yes
## 33064 -1.7 94.027 -38.3 0.900 4991.6 yes
## 33104 -1.7 94.027 -38.3 0.905 4991.6 yes
## 33109 -1.7 94.027 -38.3 0.905 4991.6 yes
## 33115 -1.7 94.027 -38.3 0.905 4991.6 yes
## 33134 -1.7 94.027 -38.3 0.904 4991.6 yes
## 33138 -1.7 94.027 -38.3 0.904 4991.6 yes
## 33142 -1.7 94.027 -38.3 0.904 4991.6 yes
## 33151 -1.7 94.027 -38.3 0.903 4991.6 yes
## 33164 -1.7 94.027 -38.3 0.899 4991.6 yes
## 33189 -1.7 94.027 -38.3 0.896 4991.6 yes
## 33214 -1.7 94.027 -38.3 0.886 4991.6 yes
## 33215 -1.7 94.027 -38.3 0.886 4991.6 yes
## 33216 -1.7 94.027 -38.3 0.886 4991.6 yes
## 33223 -1.1 94.199 -37.5 0.886 4963.6 yes
## 33241 -1.1 94.199 -37.5 0.886 4963.6 yes
## 33257 -1.1 94.199 -37.5 0.884 4963.6 yes
## 33295 -1.1 94.199 -37.5 0.881 4963.6 yes
## 33296 -1.1 94.199 -37.5 0.881 4963.6 yes
## 33298 -1.1 94.199 -37.5 0.881 4963.6 yes
## 33304 -1.1 94.199 -37.5 0.881 4963.6 yes
## 33333 -1.1 94.199 -37.5 0.879 4963.6 yes
## 33335 -1.1 94.199 -37.5 0.878 4963.6 yes
## 33346 -1.1 94.199 -37.5 0.879 4963.6 yes
## 33356 -1.1 94.199 -37.5 0.877 4963.6 yes
## 33368 -1.1 94.199 -37.5 0.876 4963.6 yes
## 33372 -1.1 94.199 -37.5 0.876 4963.6 yes
## 33383 -1.1 94.199 -37.5 0.879 4963.6 yes
## 33390 -1.1 94.199 -37.5 0.879 4963.6 yes
## 33504 -1.1 94.601 -49.5 0.972 4963.6 yes
## 33511 -1.1 94.601 -49.5 0.972 4963.6 yes
## 33519 -1.1 94.601 -49.5 0.977 4963.6 yes
## 33533 -1.1 94.601 -49.5 0.982 4963.6 yes
## 33539 -1.1 94.601 -49.5 0.985 4963.6 yes
## 33540 -1.1 94.601 -49.5 0.985 4963.6 yes
## 33560 -1.1 94.601 -49.5 0.993 4963.6 yes
## 33565 -1.1 94.601 -49.5 1.000 4963.6 yes
## 33571 -1.1 94.601 -49.5 1.000 4963.6 yes
## 33583 -1.1 94.601 -49.5 1.016 4963.6 yes
## 33590 -1.1 94.601 -49.5 1.025 4963.6 yes
## 33608 -1.1 94.601 -49.5 1.029 4963.6 yes
## 33627 -1.1 94.601 -49.5 1.043 4963.6 yes
## 33633 -1.1 94.601 -49.5 1.043 4963.6 yes
## 33670 -1.1 94.767 -50.8 1.049 4963.6 yes
## 33733 -1.1 94.767 -50.8 1.031 4963.6 yes
## 33737 -1.1 94.767 -50.8 1.031 4963.6 yes
## 33742 -1.1 94.767 -50.8 1.028 4963.6 yes
# => yesは日数が短い人の割合が大きい
# 以前のキャンペーン結果
#plot_ly(x = bank_marketing_train_job_retired$poutcome, type="histogram", color = bank_marketing_train_job_retired$y)
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$poutcome, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$poutcome, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$poutcome)/num_retired_yes
## failure nonexistent success
## 0.1424581 0.5837989 0.2737430
summary(bank_marketing_train_job_retired_n$poutcome)/num_retired_no
## failure nonexistent success
## 0.12476723 0.83985102 0.03538175
# => yesはfailure, successが多い
# 以前のキャンペーンの接触回数
pl_yes <- plot_ly(x = bank_marketing_train_job_retired_y$previous, type="histogram", name = "yes")
pl_no <- plot_ly(x = bank_marketing_train_job_retired_n$previous, type="histogram", name = "no")
subplot(pl_yes, pl_no)
# 割合をみてみる
summary(bank_marketing_train_job_retired_y$previous)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.6508 1.0000 4.0000
summary(bank_marketing_train_job_retired_n$previous)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.2086 0.0000 4.0000
# => yesは平均値が大きい(yes:0.65, no:0.20)しかし、この説明変数がどれだけ有効なのかは想像つかない
lr3<-glm(y~.-day_of_week-duration-campaign,
data=bank_marketing_train, family="binomial")
summary(lr3)
##
## Call:
## glm(formula = y ~ . - day_of_week - duration - campaign, family = "binomial",
## data = bank_marketing_train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9013 -0.4129 -0.3240 -0.2778 2.9504
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -9.508e+01 1.600e+01 -5.944 2.79e-09 ***
## age 1.383e-03 2.318e-03 0.597 0.55060
## jobblue-collar -1.963e-01 7.484e-02 -2.623 0.00871 **
## jobentrepreneur -8.459e-02 1.153e-01 -0.734 0.46322
## jobhousemaid -1.070e-01 1.389e-01 -0.770 0.44109
## jobmanagement -7.113e-02 8.211e-02 -0.866 0.38631
## jobretired 2.807e-01 1.028e-01 2.731 0.00632 **
## jobself-employed -1.509e-01 1.146e-01 -1.317 0.18772
## jobservices -1.568e-01 8.150e-02 -1.925 0.05428 .
## jobstudent 2.415e-01 1.088e-01 2.219 0.02649 *
## jobtechnician -2.015e-02 6.753e-02 -0.298 0.76540
## jobunemployed -5.372e-02 1.223e-01 -0.439 0.66040
## jobunknown -2.881e-01 2.362e-01 -1.219 0.22267
## maritalmarried 5.616e-02 6.552e-02 0.857 0.39137
## maritalsingle 1.196e-01 7.447e-02 1.605 0.10840
## maritalunknown 6.134e-02 4.103e-01 0.150 0.88115
## educationbasic.6y 4.597e-02 1.151e-01 0.399 0.68959
## educationbasic.9y -5.598e-02 8.954e-02 -0.625 0.53181
## educationhigh.school -1.073e-02 8.735e-02 -0.123 0.90226
## educationilliterate 1.192e+00 6.560e-01 1.817 0.06914 .
## educationprofessional.course 4.639e-02 9.657e-02 0.480 0.63096
## educationuniversity.degree 1.052e-01 8.729e-02 1.206 0.22791
## educationunknown 2.121e-01 1.133e-01 1.872 0.06122 .
## defaultunknown -3.176e-01 6.312e-02 -5.032 4.86e-07 ***
## defaultyes -7.644e+00 8.447e+01 -0.090 0.92790
## housingunknown -3.040e-02 1.279e-01 -0.238 0.81208
## housingyes -4.811e-02 3.933e-02 -1.223 0.22125
## loanunknown NA NA NA NA
## loanyes -8.754e-02 5.471e-02 -1.600 0.10960
## contacttelephone -8.601e-01 6.075e-02 -14.156 < 2e-16 ***
## pdays -1.176e-03 2.207e-04 -5.326 1.00e-07 ***
## previous -5.077e-02 6.156e-02 -0.825 0.40954
## poutcomenonexistent 5.005e-01 9.467e-02 5.287 1.24e-07 ***
## poutcomesuccess 6.782e-01 2.158e-01 3.143 0.00167 **
## emp.var.rate -7.105e-01 6.552e-02 -10.844 < 2e-16 ***
## cons.price.idx 1.099e+00 1.043e-01 10.534 < 2e-16 ***
## cons.conf.idx 3.879e-02 6.027e-03 6.436 1.23e-10 ***
## euribor3m 6.074e-02 8.249e-02 0.736 0.46148
## nr.employed -1.506e-03 1.451e-03 -1.038 0.29949
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 23735 on 33743 degrees of freedom
## Residual deviance: 19114 on 33706 degrees of freedom
## AIC: 19190
##
## Number of Fisher Scoring iterations: 9
## step関数
lr4 <- step(lr3)
## Start: AIC=19189.68
## y ~ (age + job + marital + education + default + housing + loan +
## contact + day_of_week + duration + campaign + pdays + previous +
## poutcome + emp.var.rate + cons.price.idx + cons.conf.idx +
## euribor3m + nr.employed) - day_of_week - duration - campaign
##
## Df Deviance AIC
## - marital 3 19117 19187
## - age 1 19114 19188
## - euribor3m 1 19114 19188
## - previous 1 19114 19188
## - nr.employed 1 19115 19189
## - education 7 19127 19189
## - housing 1 19115 19189
## <none> 19114 19190
## - loan 1 19116 19190
## - job 11 19150 19204
## - default 2 19141 19213
## - pdays 1 19142 19216
## - poutcome 2 19147 19219
## - cons.conf.idx 1 19155 19229
## - cons.price.idx 1 19221 19295
## - emp.var.rate 1 19230 19304
## - contact 1 19327 19401
##
## Step: AIC=19186.58
## y ~ age + job + education + default + housing + loan + contact +
## pdays + previous + poutcome + emp.var.rate + cons.price.idx +
## cons.conf.idx + euribor3m + nr.employed
##
## Df Deviance AIC
## - age 1 19117 19185
## - euribor3m 1 19117 19185
## - previous 1 19117 19185
## - nr.employed 1 19118 19186
## - housing 1 19118 19186
## - education 7 19130 19186
## <none> 19117 19187
## - loan 1 19119 19187
## - job 11 19158 19206
## - default 2 19143 19209
## - pdays 1 19145 19213
## - poutcome 2 19150 19216
## - cons.conf.idx 1 19159 19227
## - cons.price.idx 1 19225 19293
## - emp.var.rate 1 19234 19302
## - contact 1 19331 19399
##
## Step: AIC=19184.58
## y ~ job + education + default + housing + loan + contact + pdays +
## previous + poutcome + emp.var.rate + cons.price.idx + cons.conf.idx +
## euribor3m + nr.employed
##
## Df Deviance AIC
## - euribor3m 1 19117 19183
## - previous 1 19117 19183
## - nr.employed 1 19118 19184
## - housing 1 19118 19184
## - education 7 19130 19184
## <none> 19117 19185
## - loan 1 19119 19185
## - job 11 19160 19206
## - default 2 19144 19208
## - pdays 1 19145 19211
## - poutcome 2 19150 19214
## - cons.conf.idx 1 19159 19225
## - cons.price.idx 1 19225 19291
## - emp.var.rate 1 19234 19300
## - contact 1 19331 19397
##
## Step: AIC=19183.09
## y ~ job + education + default + housing + loan + contact + pdays +
## previous + poutcome + emp.var.rate + cons.price.idx + cons.conf.idx +
## nr.employed
##
## Df Deviance AIC
## - nr.employed 1 19118 19182
## - previous 1 19118 19182
## - housing 1 19119 19183
## - education 7 19131 19183
## <none> 19117 19183
## - loan 1 19120 19184
## - job 11 19160 19204
## - default 2 19145 19207
## - pdays 1 19145 19209
## - poutcome 2 19151 19213
## - cons.conf.idx 1 19227 19291
## - emp.var.rate 1 19248 19312
## - cons.price.idx 1 19256 19320
## - contact 1 19332 19396
##
## Step: AIC=19181.66
## y ~ job + education + default + housing + loan + contact + pdays +
## previous + poutcome + emp.var.rate + cons.price.idx + cons.conf.idx
##
## Df Deviance AIC
## - previous 1 19118 19180
## - housing 1 19119 19181
## - education 7 19132 19182
## <none> 19118 19182
## - loan 1 19120 19182
## - job 11 19161 19203
## - default 2 19145 19205
## - pdays 1 19146 19208
## - poutcome 2 19151 19211
## - cons.conf.idx 1 19253 19315
## - contact 1 19362 19424
## - cons.price.idx 1 19642 19704
## - emp.var.rate 1 20622 20684
##
## Step: AIC=19180.23
## y ~ job + education + default + housing + loan + contact + pdays +
## poutcome + emp.var.rate + cons.price.idx + cons.conf.idx
##
## Df Deviance AIC
## - housing 1 19120 19180
## - education 7 19132 19180
## <none> 19118 19180
## - loan 1 19121 19181
## - job 11 19162 19202
## - default 2 19146 19204
## - pdays 1 19147 19207
## - poutcome 2 19218 19276
## - cons.conf.idx 1 19253 19313
## - contact 1 19363 19423
## - cons.price.idx 1 19667 19727
## - emp.var.rate 1 20653 20713
##
## Step: AIC=19179.66
## y ~ job + education + default + loan + contact + pdays + poutcome +
## emp.var.rate + cons.price.idx + cons.conf.idx
##
## Df Deviance AIC
## - loan 2 19123 19179
## - education 7 19133 19179
## <none> 19120 19180
## - job 11 19163 19201
## - default 2 19147 19203
## - pdays 1 19149 19207
## - poutcome 2 19219 19275
## - cons.conf.idx 1 19256 19314
## - contact 1 19363 19421
## - cons.price.idx 1 19669 19727
## - emp.var.rate 1 20654 20712
##
## Step: AIC=19178.48
## y ~ job + education + default + contact + pdays + poutcome +
## emp.var.rate + cons.price.idx + cons.conf.idx
##
## Df Deviance AIC
## - education 7 19136 19178
## <none> 19123 19179
## - job 11 19166 19200
## - default 2 19150 19202
## - pdays 1 19151 19205
## - poutcome 2 19222 19274
## - cons.conf.idx 1 19259 19313
## - contact 1 19366 19420
## - cons.price.idx 1 19672 19726
## - emp.var.rate 1 20658 20712
##
## Step: AIC=19178.22
## y ~ job + default + contact + pdays + poutcome + emp.var.rate +
## cons.price.idx + cons.conf.idx
##
## Df Deviance AIC
## <none> 19136 19178
## - default 2 19164 19202
## - pdays 1 19165 19205
## - job 11 19191 19211
## - poutcome 2 19237 19275
## - cons.conf.idx 1 19280 19320
## - contact 1 19385 19425
## - cons.price.idx 1 19693 19733
## - emp.var.rate 1 20681 20721
AIC(lr4)
## [1] 19178.22
summary(lr4)
##
## Call:
## glm(formula = y ~ job + default + contact + pdays + poutcome +
## emp.var.rate + cons.price.idx + cons.conf.idx, family = "binomial",
## data = bank_marketing_train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9100 -0.4162 -0.3238 -0.2797 2.9583
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.110e+02 4.650e+00 -23.863 < 2e-16 ***
## jobblue-collar -2.618e-01 6.097e-02 -4.295 1.75e-05 ***
## jobentrepreneur -9.485e-02 1.138e-01 -0.834 0.404516
## jobhousemaid -1.419e-01 1.321e-01 -1.074 0.282787
## jobmanagement -4.861e-02 8.023e-02 -0.606 0.544562
## jobretired 2.660e-01 8.171e-02 3.255 0.001133 **
## jobself-employed -1.380e-01 1.136e-01 -1.214 0.224663
## jobservices -2.120e-01 7.743e-02 -2.738 0.006180 **
## jobstudent 2.540e-01 9.971e-02 2.547 0.010855 *
## jobtechnician -2.225e-02 6.004e-02 -0.371 0.710965
## jobunemployed -9.283e-02 1.206e-01 -0.769 0.441599
## jobunknown -2.458e-01 2.320e-01 -1.060 0.289202
## defaultunknown -3.172e-01 6.211e-02 -5.106 3.28e-07 ***
## defaultyes -7.640e+00 8.448e+01 -0.090 0.927943
## contacttelephone -8.798e-01 5.733e-02 -15.347 < 2e-16 ***
## pdays -1.109e-03 2.059e-04 -5.386 7.22e-08 ***
## poutcomenonexistent 5.607e-01 6.150e-02 9.116 < 2e-16 ***
## poutcomesuccess 7.352e-01 2.071e-01 3.551 0.000384 ***
## emp.var.rate -7.352e-01 1.897e-02 -38.753 < 2e-16 ***
## cons.price.idx 1.190e+00 4.988e-02 23.869 < 2e-16 ***
## cons.conf.idx 4.441e-02 3.720e-03 11.940 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 23735 on 33743 degrees of freedom
## Residual deviance: 19136 on 33723 degrees of freedom
## AIC: 19178
##
## Number of Fisher Scoring iterations: 9
ここで、ageなどの説明変数の重要性が減ってしまうのは、 emp.var.rateなどの説明変数の影響が大きいためと思われる ペルソナを定義するのに使用した説明変数と、それ以外で重要な説明変数を用いて モデリングをおこなう方針とする (emp.var.rate, cons.price.idx, cons.conf.idx を追加する)
Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.